No highly effective treatment is currently available for Alzheimer's disease. To develop a suitable in vivo model for drug screening, and to investigate the basic pathogenesis of Alzheimer's disease and related disorders, we have created models of the disorder based on expression of human tau and Abeta in the fruit fly Drosophila. Our models recapitulate key features of the human disorders. Specifically, when we express human tau in Drosophila we observe shortened lifespan, behavioral abnormalities and accumulation of abnormally phosphorylated tau protein. Unbiased forward genetic screens have revealed conserved basic pathological mechanisms, including the importance of abnormal phosphorylation of tau, oxidative stress, reactivation of cell cycle in postmitotic neurons, and abnormalities of the actin cytoskeleton. We now propose using our model of Alzheimer's disease and related tauopathies to identify drugs that can ameliorate neurotoxicity in vivo. The small size and short lifespan of fruit flies allows screening of a relatively large number of compounds in intact animals. We will test the ability of 2,000 compounds (Spectrum Collection) to reduce toxicity of tau in our model. Approximately one half of the compounds we will test are USDA approved drugs. Many of the other compounds are natural products with pre-approval clinical history. Many of the drugs we will test have proven ability to reach the brain. We anticipate that the well- characterized nature of the compounds will facilitate translation of these therapeutic compounds to testing in vertebrate animal models and to eventual use in the clinic.